Intelligent puncture path planning method and system under ultrasonic guidance

Abstract

The invention relates to an intelligent puncture path planning method and system under ultrasonic guidance, and relates to the technical field of hemodialysis rope ladder puncture, and the method comprises the following steps: an ultrasonic image acquisition step: carrying out ultrasonic scanning on a preset puncture target area of a patient to obtain ultrasonic image data, and synchronously collecting arteriovenous blood vessel flow; a puncture model construction step: constructing an arm three-dimensional model corresponding to the puncture target area of the patient according to the ultrasonic image data, and analyzing the arm three-dimensional model to determine a blood vessel model and a muscle tissue model; a risk tissue avoiding step: performing analysis according to the blood vessel model and the arteriovenous blood vessel flow to determine target blood vessels and key tissues meeting puncture conditions, and constructing a three-dimensional safe puncture space; and a puncture path generation step: generating a plurality of puncture paths meeting puncture angle constraints according to the target blood vessels and the key tissues, and screening out an optimal puncture path. The method has the effects that the dependence of puncture work on medical care experience is reduced, and the overall puncture work efficiency is improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of hemodialysis rope ladder puncture, in particular to an intelligent puncture path planning method and system under ultrasonic guidance. BACKGROUND

[0002] In medical rescue work, when treating some specific patients with weakened blood metabolism filtration function, a hemodialysis method is needed to help the patient filter blood metabolites, so that the blood metabolites are filtered and then transported back into the patient's body to maintain normal operation of the human body.

[0003] In the related art, before hemodialysis, the patient's arteriovenous fistula is usually punctured by rope ladder to complete the blood circuit connection between the patient and the filtration dialysis instrument. However, due to the different physical conditions of patients, an ultrasonic examination of the punctured arteriovenous fistula is required before the rope ladder puncture, so as to understand the blood vessel structure and body tissue of the patient, and plan the puncture path according to the ultrasonic examination result.

[0004] In view of the above related art, due to the great difference between the arteriovenous fistula and the body tissue of different patients, the medical staff relies more on the puncture experience of the medical staff when performing ultrasonic examination and planning the puncture path, which is difficult to work and is not conducive to improving the overall puncture work efficiency. SUMMARY

[0005] In order to reduce the dependence of puncture work on medical experience, so that the patient can quickly and conveniently determine the puncture path of the arteriovenous fistula before hemodialysis, and improve the overall puncture work efficiency, the application provides an intelligent puncture path planning method and system under ultrasonic guidance.

[0006] In a first aspect, the application provides an intelligent puncture path planning method under ultrasonic guidance, which adopts the following technical scheme:

[0007] An intelligent puncture path planning method under ultrasonic guidance, comprising:

[0008] An ultrasonic image acquisition step, which acquires ultrasonic image data of a patient's preset puncture target area by ultrasonic scanning of the patient's preset puncture target area according to a preset scanning strategy, and synchronously acquires the arteriovenous fistula blood flow of the preset puncture target area;

[0009] A puncture model construction step, which constructs a three-dimensional model of the patient's puncture target area according to the ultrasonic image data, and analyzes the three-dimensional model of the arm by a feature recognition algorithm to determine a blood vessel model and a muscle tissue model;

[0010] The risk organization avoidance step analyzes according to the blood vessel model and arteriovenous blood flow to determine the target blood vessel and key tissue that meet the puncture conditions, and constructs a three-dimensional safe puncture space within a preset safety distance threshold according to the identification result of the target blood vessel and the key tissue;

[0011] The puncture path generation step generates a plurality of puncture paths that meet the puncture angle constraint within the three-dimensional safe puncture space according to the target blood vessel and the key tissue, and configures a path evaluation strategy for evaluation to screen out the optimal puncture path.

[0012] By adopting the above technical solutions, by combining ultrasound image and arteriovenous blood flow acquisition for model construction and path evaluation, it is helpful for three-dimensional modeling and accurate path planning, improves the accuracy of target blood vessel identification, ensures reasonable path and selects the optimal scheme, so that medical staff do not need to rely on puncture work experience to construct puncture path, reduces the risk of key tissue damage, and improves puncture safety and overall work efficiency.

[0013] Optionally, the risk organization avoidance step is also configured with a risk disease identification strategy, which includes:

[0014] Based on the blood vessel model, the arteriovenous blood vessel diameter change rate is determined, and the standard arteriovenous blood vessel diameter change rate corresponding to the puncture target area in the preset blood vessel disease database is matched for comparison to obtain a blood vessel diameter change comparison result;

[0015] Based on the muscle tissue model, the muscle tissue model parameters are obtained, and the preset muscle disease model database is compared to determine a muscle model comparison result;

[0016] According to the blood vessel diameter change comparison result and the muscle model comparison result, the blood vessel position and muscle area that do not meet the preset puncture risk condition are marked to determine the puncture avoidance area;

[0017] Based on the three-dimensional safe puncture space and the puncture avoidance area, the overlapping area is divided to determine the three-dimensional safe puncture space avoiding the puncture avoidance area.

[0018] By adopting the above technical solutions, the strategy uses the comparison and analysis of the blood vessel diameter change rate and the muscle tissue parameters, and realizes effective identification of blood vessel and muscle diseases in combination with the preset disease database, which helps to accurately locate the risk area that does not meet the puncture condition, avoids puncture complications caused by blood vessel or muscle diseases, and further optimizes the reliability of the three-dimensional safe puncture space through the overlapping division of the avoidance area and the safe space, and improves the pertinence and accuracy of the risk organization avoidance.

[0019] Optionally, the puncture path generation step further includes a puncture point analysis sub-strategy:

[0020] According to the ultrasonic image parameter analysis, a skin surface tissue image of the puncture target region is obtained, and image feature recognition is performed to determine whether the skin surface is consistent with the preset skin damage feature;

[0021] The region consistent with the preset skin damage feature is marked for screening to determine a non-puncture starting point region and an available puncture starting point region;

[0022] Based on the available puncture starting point region, a related region of a puncture path is generated, and a target puncture starting point is marked according to an optimal puncture path, the target puncture starting point including a venous end starting point and an arterial end starting point.

[0023] By adopting the above technical solution, the sub-strategy analyzes the skin surface features based on the ultrasonic image, accurately identifies the damaged region, and divides the available puncture region, which helps to avoid puncture at the damaged skin position, reduces the risk of infection or bleeding after puncture, and accurately positions the arterial end and venous end starting points by associating the available region with the optimal path marking starting point, thereby improving the rationality of the puncture starting point selection and the success rate of the puncture operation.

[0024] Optionally, a puncture angle constraint optimization strategy of the target puncture starting point is further configured:

[0025] The puncture angle constraint strategy calculates an optimal puncture angle according to a preset puncture constraint angle model;

[0026] The puncture constraint angle model is calculated by the following formula:

[0027]

[0028] wherein, θ optimal is the optimal puncture angle with the skin surface, d is the puncture target blood vessel depth measured by ultrasound, p m is a preset muscle texture resistance coefficient of the muscle region where the target puncture starting point is located when the puncture needle is inserted, f is the angle between the blood vessel direction measured by ultrasound and the skin surface, t is the subcutaneous tissue thickness measured by ultrasound, b is the muscle tension adjustment coefficient, and T m is a preset muscle tension index.

[0029] By adopting the above technical solution, the optimal puncture angle is calculated by using the model formula, and the blood vessel depth, muscle texture resistance, blood vessel direction and other multi-dimensional parameters are combined to realize the quantitative optimization of the puncture angle, which helps to avoid the problems of excessive puncture resistance or target point deviation caused by improper angle selection, improves the adaptability of the puncture angle to the individual tissue characteristics of the patient, reduces the tissue damage during puncture, and improves the accuracy and comfort of the puncture operation.

[0030] Optionally, the strategy further comprises a rotation puncture starting point construction strategy:

[0031] According to the puncture starting point and the blood vessel model, analysis is performed to determine the proximal end of the blood vessel model as an extension direction;

[0032] According to the extension direction and a preset reference interval distance, a plurality of puncture points are marked as rotation puncture starting points;

[0033] According to the rotation puncture starting points and a preset puncture point analysis sub-strategy, analysis is performed to generate a corresponding puncture angle constraint for each rotation puncture starting point.

[0034] By adopting the above technical solution, the strategy is based on the blood vessel extension direction and the reference interval to mark the rotation puncture points, and combines the puncture point analysis sub-strategy to generate the corresponding angle constraint, which helps to realize the ordered rotation of the puncture starting points, avoids the damage or stenosis of the blood vessel wall caused by repeated puncture of the same area, and prolongs the service life of the blood vessel access. At the same time, through the angle constraint matching, the standardization of the puncture operation of each rotation point is guaranteed, and the risk of puncture complications is reduced.

[0035] Optionally, the path evaluation strategy comprises:

[0036] A preset puncture point risk evaluation model is used for calculation and analysis to determine the puncture risk weight value of the puncture point corresponding blood vessel when punctured according to the puncture path;

[0037] Based on the puncture risk weight value, the plurality of puncture paths are arranged in descending order to determine the puncture path with the minimum puncture risk weight value and mark it as the optimal puncture path.

[0038] By adopting the above technical solution, the strategy uses the puncture point risk evaluation model to quantitatively analyze the puncture risk, and selects the optimal path based on the risk weight value, which helps to realize the risk level differentiated evaluation of the puncture path, accurately locate the low-risk path, avoid the damage of blood vessels or tissues caused by improper path selection, and improve the scientificity and safety of the puncture path selection.

[0039] Optionally, the puncture point risk evaluation model is calculated by the following formula:

[0040]

[0041] wherein, R i is the comprehensive risk index of the i i th puncture point, K is the access type correction coefficient of the preset different access blood vessels, N max is the number of punctures within the last 2 weeks of the i i th puncture point, L i is the maximum length of the effective puncture segment of the blood vessel access, L i is the straight line distance between the i i th puncture point and the anastomosis.is the blood vessel diameter at the i-th puncture point, which is measured by ultrasound, V i is the blood flow velocity at the i-th puncture point, S i is the interval between the i-th puncture point and the adjacent puncture point, is a preset time attenuation coefficient, T i is the interval between the i-th puncture point and the last puncture.

[0042] By adopting the above technical scheme, the model quantitatively calculates the puncture point comprehensive risk index through the formula, integrates multiple-dimensional parameters such as path type, puncture times, and blood vessel diameter, realizes quantifiable evaluation of the risk, helps to accurately distinguish the risk difference of different puncture points, provides objective data support for optimal path screening, avoids risk misjudgment caused by subjective judgment, and improves the accuracy and reliability of puncture point risk evaluation.

[0043] Optionally, it further includes a puncture parameter learning step:

[0044] The tip position and needle body posture of the physical puncture needle in the three-dimensional image model are acquired in real time through electromagnetic tracking or image recognition technology;

[0045] The real-time trajectory of the physical puncture needle is compared with the optimal puncture path, and when it deviates from the path or approaches the risk tissue boundary beyond a preset safety threshold, a visual or audible early warning signal is triggered and a puncture path deviation data packet is generated;

[0046] The puncture path deviation data packet is input into a preset automatic puncture path planning intelligent agent for path deviation analysis to generate a puncture deviation feedback result.

[0047] By adopting the above technical scheme, this step uses electromagnetic tracking or image recognition to monitor the puncture needle trajectory in real time, combines early warning and deviation analysis feedback, realizes dynamic correction of the puncture process, helps to discover and prompt path deviation or risk approach problems in time, avoids tissue damage caused by operation deviation; at the same time, the feedback result is generated through intelligent agent analysis, which provides data support for subsequent puncture parameter optimization, and improves the fault tolerance and accuracy of puncture operation.

[0048] Optionally, the scanning strategy includes:

[0049] The patient's preset puncture target area is analyzed to determine the puncture limb type and match the moving speed of the ultrasound scanner corresponding to the puncture limb type in the preset ultrasound scanning database;

[0050] The preset puncture target area is scanned based on the moving speed of the ultrasound scanner, and the real-time fit degree of the preset ultrasound scanner is analyzed to keep the real-time fit degree within a preset reference fit degree range;

[0051] When the moving speed of the ultrasound scanner and the tightness do not meet the preset scanning conditions, a moving speed prompt instruction or a tightness adjustment prompt instruction is issued.

[0052] By adopting the technical solution, the strategy matches the scanning speed based on the limb type, combines the tightness monitoring and adjustment prompt, realizes the standardized operation of ultrasound scanning, helps to avoid image blur caused by too fast scanning speed or image artifacts caused by insufficient tightness, guarantees the clarity and integrity of the ultrasound image data, corrects the scanning deviation in time through the prompt instruction, provides reliable basic data support for subsequent model construction and path planning, and improves the accuracy of the overall puncture planning.

[0053] In a second aspect, the application provides an intelligent puncture path planning system under ultrasound guidance, which adopts the following technical solution:

[0054] An intelligent puncture path planning system under ultrasound guidance, which applies the method described above, comprises:

[0055] An ultrasound image acquisition module acquires ultrasound image data of a patient's preset puncture target area by performing ultrasound scanning on the preset puncture target area according to a preset scanning strategy, and synchronously acquires the arterial and venous blood flow of the preset puncture target area;

[0056] A puncture model construction module constructs a three-dimensional model of the patient's puncture target area according to the ultrasound image data, and analyzes the three-dimensional model by a feature recognition algorithm to determine a blood vessel model and a muscle tissue model;

[0057] A risk tissue avoidance module analyzes the blood vessel model and the arterial and venous blood flow to determine a target blood vessel and key tissues that meet the puncture conditions, and constructs a three-dimensional safe puncture space within a preset safety distance threshold according to the recognition results of the target blood vessel and the key tissues;

[0058] A puncture path generation module generates a plurality of puncture paths in the three-dimensional safe puncture space that meet the puncture angle constraints according to the target blood vessel and the key tissues, and configures a path evaluation strategy to evaluate and select an optimal puncture path.

[0059] By adopting the technical solution, three-dimensional modeling and accurate path planning are realized through model construction, risk avoidance and path evaluation by combining ultrasound image acquisition and arterial and venous blood flow acquisition, the target blood vessel recognition accuracy is improved, the path is guaranteed to be reasonable and the optimal solution is selected, the risk of key tissue damage is reduced, and the puncture safety and efficiency are improved.

[0060] In summary, the application has at least one of the following beneficial technical effects:

[0061] 1. By combining ultrasound images with arteriovenous blood flow acquisition, model construction, risk avoidance, and path assessment, it helps to three-dimensional modeling and precise path planning, improves target vessel identification accuracy, ensures reasonable path and selects the optimal solution, so that medical staff do not need to rely on puncture experience to construct puncture path, reduces the risk of key tissue damage, and improves puncture safety and overall work efficiency;

[0062] 2. By comparing the blood vessel diameter change rate with the muscle tissue parameter, combined with the preset disease database to identify the risk of blood vessels and muscle diseases, accurately locate the risk area to avoid puncture complications, optimize the three-dimensional safe puncture space, and improve the targeting and accuracy of risk tissue avoidance;

[0063] 3. Based on the analysis of skin features in ultrasound images, identify damaged areas and divide available puncture areas to reduce the risk of infection or bleeding, associate available area markers to start point to achieve accurate positioning of arteriovenous starting point, and improve the rationality of puncture starting point and operation success rate. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 is a method flowchart of steps S100 to S400 in this application.

[0065] Figure 2 is a method flowchart of steps S301 to S304 in this application.

[0066] Figure 3 is a method flowchart of steps S401 to S403 in this application.

[0067] Figure 4 is a method flowchart of steps S404 to S406 in this application.

[0068] Figure 5 is a method flowchart of steps S407 to S408 in this application.

[0069] Figure 6 is a method flowchart of steps S500 to S502 in this application.

[0070] Figure 7 is a method flowchart of steps S101 to S103 in this application. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical scheme and advantages of the present application clearer, the following will combine the drawings of the specification Figures 1-7 and examples to further illustrate the present application. It should be understood that the specific examples described herein are only used to explain the present application and not to limit the present application.

[0072] The embodiments of the present application will be further described in detail below in conjunction with the drawings of the specification.

[0073] The embodiment of the application discloses an intelligent puncture path planning method under ultrasonic guidance, which combines ultrasonic images and dynamic and static blood vessel flow acquisition to perform model construction, risk avoidance and path evaluation, and performs three-dimensional modeling of blood vessels and muscle tissues and puncture path planning, so that the target blood vessel recognition accuracy is improved, medical staff do not need to rely on puncture work experience to construct a puncture path, the risk of key tissue damage is reduced, and the puncture safety and overall work efficiency are improved.

[0074] Reference Figure 1 The method flow of the intelligent puncture path planning method under ultrasonic guidance comprises the following steps:

[0075] The ultrasonic image acquisition step S100: The preset puncture target region of the patient is scanned by a preset scanning strategy to obtain ultrasonic image data, and the dynamic and static blood vessel flow of the preset puncture target region is synchronously acquired.

[0076] The preset puncture target region refers to the region where the limb position of the patient for hemodialysis puncture is located, and is usually the dynamic and static vein fistula region of the arm, such as the vicinity of the forearm radial artery-head vein fistula, which is the core target range of the puncture operation; the ultrasonic image data: a continuous image sequence obtained by scanning the preset puncture target region by an ultrasonic device, which can reflect the shape and positional relationship of the skin, subcutaneous tissue, blood vessels, muscles and the like; the dynamic and static blood vessel flow refers to the flow velocity and flow of blood in the target region, which can be measured by ultrasonic Doppler technology and is used to evaluate the patency of the blood vessel; the preset scanning strategy: a scanning rule set to ensure the image quality and data integrity.

[0077] In specific implementation, the system first determines the preset puncture target region of the patient, such as the left forearm fistula region, and then controls the ultrasonic probe to scan the region according to the preset scanning strategy, for example, moves the probe at a constant speed from the fistula anastomosis to the distal end, synchronously generates continuous ultrasonic image data; at the same time, the flow data of the dynamic and static blood vessels in the region is synchronously acquired in real time by the ultrasonic Doppler mode, such as the arterial end flow ≥500ml / min as an ideal state, and the image data and the flow data are stored in association.

[0078] The role of this step is to provide an original data basis for subsequent puncture planning: the ultrasonic image data can directly reflect the anatomical structure of the target region, and the dynamic and static blood vessel flow can evaluate the functional state of the blood vessel. When the flow is too low, it may indicate that the blood vessel is stenotic and needs to be focused on, for example, if the scanning finds that the venous blood vessel flow of the fistula region of a certain patient is 300ml / min, which is lower than the ideal value, the system will associate the data with the shape of the blood vessel in the ultrasonic image to provide a basis for subsequent determination of the target blood vessel. It should be noted that the specific content of the preset scanning strategy will be further described in the subsequent steps.

[0079] Puncture model construction step S200: constructing a three-dimensional model of the patient's arm corresponding to the puncture target region according to the ultrasound image data, and analyzing the three-dimensional model of the arm by a feature recognition algorithm to determine a blood vessel model and a muscle tissue model;

[0080] Three-dimensional model of the arm: a three-dimensional model of the preset puncture target region reconstructed based on the ultrasound image data, which can visually display the spatial positional relationship of the skin, subcutaneous tissue, blood vessels, and muscles. Blood vessel model: a blood vessel structure sub-model extracted from the three-dimensional model of the arm, containing details such as the direction, diameter, length, and branches of the blood vessels, such as the main stem and branches of the internal fistula blood vessels. Muscle tissue model: a muscle structure sub-model extracted from the three-dimensional model of the arm, containing information such as the texture direction, thickness, and relative position to blood vessels of the muscle. Feature recognition algorithm: an algorithm for automatically recognizing and extracting target structures such as blood vessels and muscles from a three-dimensional model, the specific logic of which will be further explained in subsequent steps.

[0081] In specific implementation, the system first performs three-dimensional reconstruction on the ultrasound image data obtained in step S100. For example, by matching the spatial coordinates of consecutive images, a three-dimensional grid model of the target region of the arm is generated; then the feature recognition algorithm is started to analyze the structural features in the three-dimensional model: blood vessel structures are recognized by the tubular shape and blood flow signals of the blood vessels, such as combined with the arterial and venous blood flow data to generate a blood vessel model; muscle tissue is recognized by the texture distribution and density characteristics of the muscle to generate a muscle tissue model, and the spatial positional relationship between the two is labeled, such as a certain segment of blood vessels being located below the biceps brachii muscle.

[0082] The purpose of this step is to convert two-dimensional ultrasound images into three-dimensional structured models, achieving visualization and structured expression of the puncture target region. Compared with traditional two-dimensional images, three-dimensional models can more clearly display the spatial relationship between blood vessels and muscles, providing intuitive basis for subsequent determination of safe puncture range. For example, the diameter and length of the internal fistula blood vessels can be directly measured by the blood vessel model, providing data support for determining whether the puncture conditions are met. It should be noted that the specific content of the feature recognition algorithm will be further explained in subsequent steps.

[0083] Risk tissue avoidance step S300: analyzing the blood vessel model and arterial and venous blood flow to determine target blood vessels and key tissues that meet the puncture conditions, and constructing a three-dimensional safe puncture space within a preset safe distance threshold according to the identification results of the target blood vessels and key tissues;

[0084] Target vessel: refers to the vessel that meets the puncture conditions, which is the target vessel for puncture operation, such as a vein segment with a tube diameter of ≥2.5 mm and a flow rate of ≥500 ml / min in an internal fistula. The conditions referred to are that the tube diameter meets the standard, the flow rate is normal, and there is no obvious stenosis. Key tissues refer to the risk tissues adjacent to the target vessel, including nerves, tendons, bones, and abnormal blood vessel segments. The three-dimensional safe puncture space refers to a three-dimensional area centered on the target vessel, avoiding key tissues and meeting the preset safety distance, which is the safe range for the puncture needle to pass through. The preset safety distance threshold refers to the minimum distance set to avoid puncture damage to key tissues, such as a safety distance of ≥0.5 cm from the nerve.

[0085] In specific implementation, the system first analyzes the tube diameter, direction, and integrity of the blood vessel based on the blood vessel model, evaluates the blood vessel function in combination with the arterial and venous blood flow, such as screening a vein segment with a tube diameter of 2.8 mm, a flow rate of 550 ml / min, and a straight direction as the target vessel; at the same time, the key tissues around the target vessel are identified through the muscle tissue model, such as the ulnar nerve 0.3 cm away from the vein segment; then, a region that does not contain key tissues is divided in the three-dimensional model according to the preset safety distance threshold, such as maintaining a distance of ≥0.5 cm from the nerve, with the target vessel as the core, that is, the three-dimensional safe puncture space.

[0086] The role of this step is to define the safe operation boundary. By accurately identifying the target vessel and key tissues, and combining safety distance constraints to construct a safe space, the risk of puncture needle damage to risk tissues is avoided from the root. For example, the three-dimensional safe space around the above-mentioned target vessel will automatically avoid the ulnar nerve area, ensuring that the subsequent generated puncture path is within the safe range. It should be noted that the specific rules of key tissue identification and safe space construction will be further explained in subsequent steps.

[0087] Puncture path generation step S400: generating multiple puncture paths that meet the puncture angle constraint in the three-dimensional safe puncture space according to the target vessel and key tissues, and configuring a path evaluation strategy for evaluation to screen the optimal puncture path.

[0088] Puncture angle constraint refers to the range of the angle between the puncture needle and the skin surface, which is commonly used in clinical practice at 30°-45°, and needs to be determined in combination with factors such as blood vessel depth and subcutaneous tissue thickness. The candidate puncture path refers to a continuous path from the skin surface to the target vessel, which needs to be completely located in the three-dimensional safe puncture space and meet the angle constraint. The path evaluation strategy is a rule for comprehensive evaluation of the candidate path, such as evaluating the path length and distance from the risk tissue, and the specific content will be further explained in subsequent steps. The optimal puncture path represents the path with the highest score after evaluation, which usually has the characteristics of low operation difficulty and small damage risk.

[0089] In specific implementation, the system takes the three-dimensional safe puncture space as the range, generates multiple candidate paths that meet the puncture angle constraint from the possible puncture point on the skin surface to the target blood vessel, for example, 5 paths with an angle in the range of 35° to 40° for a certain target blood vessel, each of which avoids key tissues; then scores the paths through a path evaluation strategy, for example, the shorter the path and the farther the distance from the risk tissue, the higher the score, and finally filters out the path with the highest score as the optimal puncture path.

[0090] The purpose of this step is to provide accurate operation guidance: by generating multiple candidate paths and selecting the optimal scheme, the subjectivity of relying on manual experience to select the path is avoided, and it is ensured that the puncture path is both safe and easy to operate. For example, the optimal path may select the point closest to the target blood vessel directly above the skin surface to puncture at the shortest distance and appropriate angle, reducing the difficulty of operation. It should be noted that the specific content of the path evaluation strategy will be further explained in subsequent steps.

[0091] Reference Figure 2 The risk tissue avoidance step is also configured with a risk disease identification strategy, including:

[0092] Step S301: identification analysis based on the blood vessel model to determine the arteriovenous blood vessel diameter change rate, and match the standard blood vessel diameter change rate corresponding to the puncture target area in the preset blood vessel disease database to obtain a blood vessel diameter change comparison result;

[0093] Arteriovenous blood vessel diameter change rate: refers to the diameter change proportion of the blood vessel along the length direction, such as the diameter increase / decrease amplitude per centimeter of the blood vessel, which can be calculated through the continuously measured pipe diameter data in the blood vessel model, and is used to judge whether the blood vessel has abnormalities such as stenosis, dilation, etc. such as aneurysm;

[0094] Pre-set blood vessel disease database: a database storing blood vessel morphological feature data corresponding to different blood vessel diseases, containing standard parameters of the puncture target area;

[0095] Standard blood vessel diameter change rate: the diameter change rate benchmark value of the normal blood vessel recorded in the preset blood vessel disease database, for example, the normal internal fistula blood vessel diameter change rate is usually ≤5% / cm;

[0096] Blood vessel diameter change comparison result: the comparison result of the arteriovenous blood vessel diameter change rate and the standard blood vessel diameter change rate, for example, the change rate exceeds the standard value by 20% or meets the standard range, which is used to judge whether the blood vessel has disease risk.

[0097] In specific implementation, the system extracts the continuous tube diameter data of the target blood vessel from the blood vessel model, such as measuring the diameter every 0.5 cm, and calculates the diameter change rate; then, the system retrieves the standard blood vessel diameter change rate corresponding to the fistula region in the preset blood vessel disease database, compares the calculated value with the standard value, and obtains the comparison result that the change rate 33%>5%, which exceeds the standard range.

[0098] This step serves to identify the risk of blood vessel disease through quantitative comparison, and provides a basis for subsequent puncture avoidance. For example, if the comparison finds that the diameter change rate of a certain segment of blood vessel abnormally increases, it may indicate the risk of blood vessel stenosis or aneurysm, and the system will use this result as a reference for subsequent marking of avoidance regions.

[0099] Step S302: identifying and analyzing based on the muscle tissue model to obtain muscle tissue model parameters, and comparing the parameters with a preset muscle disease model database to determine a muscle tissue model comparison result;

[0100] The muscle tissue model parameters refer to the quantitative features extracted from the muscle tissue model, including muscle echo intensity, texture continuity, and relative displacement degree with blood vessels, which are core indicators for judging the health status of muscle;

[0101] The preset muscle disease model database refers to a database that stores the parameter thresholds of common muscle diseases in the puncture target region, such as common muscle diseases in the arm, such as intermuscular edema, fibrosis, and inflammation, including normal muscle parameter ranges, such as echo intensity -50 to -40 dB, and disease state parameters, such as edematous muscle echo intensity <-55 dB;

[0102] The muscle tissue model comparison result refers to the comparison conclusion of the muscle tissue model parameters with the normal parameter range in the database, such as echo intensity -60 dB <-55 dB, which meets the edema characteristics, or texture continuity 90%>80%, which meets the normal characteristics. By calculating the parameter deviation degree, such as the percentage of echo intensity deviation from the normal range, it is determined whether there is a disease risk and the risk type.

[0103] In specific implementation, the system extracts parameters from the muscle tissue model, such as target region muscle echo intensity -58 dB and texture continuity 60%; calls the standard parameters of arm muscle in the preset muscle disease model database, compares the extracted parameters with the standard parameters, and obtains the comparison result that the echo intensity is lower than the normal range, the texture continuity is insufficient, and it meets the characteristics of intermuscular edema.

[0104] The purpose of this step is to identify the risk of muscle lesions through parameter quantification comparison, and to provide the basis for puncture avoidance. For example, if the comparison finds that the muscle is edematous, the tissue is fragile during puncture, which may cause bleeding. This result will directly affect the subsequent delineation of the avoidance area. It should be noted that the specific algorithm for extracting muscle parameters and the matching rules of the database will be further explained in the subsequent steps.

[0105] Step S303: Marking the blood vessel positions and muscle regions that do not meet the preset puncture risk conditions according to the blood vessel diameter change comparison results and muscle model comparison results to determine the puncture avoidance area;

[0106] Preset puncture risk condition: The judgment threshold set based on the clinical puncture safety standard, including: blood vessel diameter change rate ≤10% / cm, exceeding which indicates stenosis / aneurysm risk; muscle echo intensity ≥-55dB, texture continuity ≥60%, adhesion degree ≤0.5, any parameter not meeting which indicates muscle lesion risk;

[0107] Puncture avoidance area: The prohibited puncture area marked in the three-dimensional model, displayed in red highlight, including abnormal blood vessel segments and diseased muscle areas.

[0108] In specific implementation, the system marks according to the following rules: for blood vessels, if the diameter change comparison result is a change rate of 33%>10%, mark along the length direction of the blood vessel segment, and expand the width to 0.5cm outside the blood vessel to avoid the needle being inserted obliquely; for muscles, if the muscle model comparison result is intermuscular edema combined with inflammation, mark a 1cm diameter spherical region centered on the lesion to cover the lesion and the surrounding potentially damaged tissue; finally, integrate the two types of marked areas into the puncture avoidance area, and store the three-dimensional coordinate range.

[0109] The purpose of this step is to clearly mark high-risk areas and spatially delineate puncture prohibited areas to avoid secondary damage to abnormal blood vessels and diseased muscles during puncture, reducing the risk of complications such as bleeding and infection.

[0110] Step S304: Perform overlap area division based on the three-dimensional safe puncture space and the puncture avoidance area to determine the three-dimensional safe puncture space that avoids the puncture avoidance area.

[0111] Overlap area division: Use the space intersection algorithm to calculate the overlapping part of the three-dimensional safe puncture space and the puncture avoidance area, and determine the boundary coordinates of the overlapping area by comparing the three-dimensional coordinate ranges of the two.

[0112] Three-dimensional safe puncture space that avoids the puncture avoidance area: The remaining space after excluding the overlapping area, whose boundary is determined by the original three-dimensional safe puncture space boundary and the puncture avoidance area boundary, and only contains normal blood vessels, normal muscles and non-diseased tissues.

[0113] Specifically, the system realizes the following steps: importing the three-dimensional safe puncture space and the puncture avoidance area, such as the coordinate range of x∈[8,10] cm, y∈[4,6] cm, z∈[0,3] cm, into the space calculation module; identifying the overlapping area as x∈[8,10] cm, y∈[4,6] cm, z∈[0,3] cm through the intersection algorithm; subtracting the overlapping area from the original three-dimensional safe puncture space to obtain the final three-dimensional safe puncture space, i.e., x∈[5,8)∪(10,15] cm, y∈[2,4)∪(6,8] cm, z∈[0,3] cm.

[0114] The role of this step is to optimize the effectiveness of the safe puncture space by accurately excluding the lesion risk area, to ensure that the subsequent generated puncture path only passes through normal tissues, and to further improve the safety and reliability of the puncture operation.

[0115] Reference Figure 3 The puncture path generation step further includes a puncture point analysis sub-strategy:

[0116] S401: Obtain the skin surface tissue image of the puncture target area according to the ultrasonic image parameter analysis, and perform image feature recognition to determine whether the skin surface is consistent with the preset skin damage feature;

[0117] Ultrasonic image parameters: quantifiable feature indicators in ultrasonic images, including gray value, texture roughness, edge continuity, etc., used to reflect the physical state of the skin surface.

[0118] Skin surface tissue image: an image of the skin surface layer of the puncture target area extracted through ultrasonic image parameter analysis, which can clearly show details such as skin texture, damage, and scars.

[0119] Pre-set skin damage feature: a feature standard for skin that is not suitable for puncture based on clinical data, including scars, bruises, diffuse areas, skin ulcers, etc.

[0120] Specifically, the system calls the ultrasonic image data obtained in step S100 to extract the gray value, texture roughness, and other ultrasonic image parameters; reconstructs the skin surface tissue image of the puncture target area through these parameters, such as clearly displaying the skin texture and possible scars in the forearm fistula area; then starts the image feature recognition algorithm to compare the features of the skin surface tissue image with the pre-set skin damage feature to determine whether there are damaged conditions such as scars and bruises.

[0121] Step S401 realizes accurate identification of skin damage features based on ultrasound image parameters, providing a basis for subsequent division of puncture areas. It connects with the previous risk tissue avoidance step S300, i.e. the risk tissue avoidance step has excluded the deep risk of blood vessels and muscles, and this step further screens the risk from the skin surface layer, realizing full-level risk screening of the surface layer + deep layer.

[0122] S402: Mark the area consistent with the preset skin damage features to determine the non-puncture starting point area and the available puncture starting point area;

[0123] Non-puncture starting point area: refers to the skin area marked as unsuitable for puncture, mainly including areas consistent with preset skin damage features, such as scar area, bruise area, ulcer area, etc.

[0124] Available puncture starting point area: refers to the skin area that is not marked as damaged, with intact skin and normal texture, meeting the puncture surface layer conditions, and is the basis for generating the puncture starting point in the subsequent step.

[0125] In specific implementation, according to the identification result of step S401, the system marks the area consistent with the preset skin damage features, such as a 0.8cm diameter scar area, with red highlight to determine the non-puncture starting point area; at the same time, the intact skin area 2cm outside the scar is marked as the available puncture starting point area, and its boundary range is marked with a green frame.

[0126] Step S402 completes the area division based on the identification result of S401, and avoids the subsequent puncture path starting from the damaged skin by clearly marking the non-puncture area and the available area. It connects with the feature identification result of S401, and converts the judgment of whether it is damaged into the division of whether it is available, providing a clear range basis for determining the specific puncture starting point in the next step.

[0127] S403: Based on the available puncture starting point area as the puncture path generation associated area, and according to the optimal puncture path marking to obtain the target puncture starting point, the target puncture starting point includes the venous end starting point and the arterial end starting point.

[0128] Puncture path generation associated area: refers to the skin area directly related to the generation of puncture path, i.e. the available puncture starting point area, which is the starting point range of the puncture path from the skin to the target blood vessel.

[0129] Target puncture starting point: refers to the final determined puncture needle insertion point from the available puncture starting point area, which is the starting point of the puncture path.

[0130] Venous end starting point and arterial end starting point: corresponding to the venous and arterial needle entry points in hemodialysis puncture respectively, the venous end starting point is usually close to the venous end of the internal fistula, the arterial end starting point is close to the arterial end of the internal fistula, and the two need to maintain a certain distance, usually ≥ 3 cm to avoid blood flow interference.

[0131] In specific implementation, the system sets the available puncture starting point region determined in step S402 as a puncture path generation associated region; in combination with the optimal puncture path generated in step S400, i.e. a safe path from the skin to the target vessel, the system screens a position matching the starting point of the optimal path in the associated region, for example, the optimal path needs to enter vertically from a certain point on the skin to directly reach the venous target point, so the system marks the point as the venous end starting point in the available region; similarly, the arterial end starting point is marked, and finally the target puncture starting point is formed.

[0132] Step S403 is based on the available region determined in S402, in combination with the optimal puncture path to accurately locate the starting point, realizing a closed loop of region screening → specific point determination. It takes the region division result of S402 as the basis, further focuses the available region to a specific needle entry point, and distinguishes the venous end and the arterial end, both to ensure the integrity of the starting point skin and to match the angle and direction of the optimal path, to provide clear starting point guidance for subsequent puncture operation, and to reduce the risk of path deviation caused by improper starting point selection.

[0133] In addition, a puncture angle constraint optimization strategy of the target puncture starting point is also configured:

[0134] The puncture angle constraint strategy calculates the optimal puncture angle based on a preset puncture constraint angle model;

[0135] The puncture constraint angle model is calculated by the following formula:

[0136]

[0137] Wherein, θ optimal is the optimal puncture angle with the skin surface, d is the puncture target vessel depth measured by ultrasound, ρ m is the preset muscle texture resistance coefficient of the muscle region where the target puncture starting point is located when the puncture needle is inserted, φ is the angle between the blood vessel direction measured by ultrasound and the skin surface, t is the subcutaneous tissue thickness measured by ultrasound, β is the muscle tension adjustment coefficient, and T m is the preset muscle tension index.

[0138] For example, assuming that the vessel depth d = 6 mm, the subcutaneous tissue thickness t = 3 mm, the angle between the blood vessel direction and the skin surface φ = 30°, the muscle texture resistance coefficient of the oblique muscle ρ m = 1.0, the muscle tension index T m = 0.5 in a mild tension state, and the muscle tension adjustment coefficient β = 0.2:

[0139]

[0140] Referring Figure 4 In addition, the rotation also includes rotating the puncture starting point construction strategy:

[0141] S404: Analyze according to the puncture starting point and the blood vessel model to determine the proximal end of the blood vessel model as the extension direction;

[0142] The puncture starting point refers to the target puncture starting point determined in step S403, including the venous end starting point and the arterial end starting point, which is the initial needle insertion position of the puncture operation.

[0143] The blood vessel model refers to the sub-model containing information such as blood vessel direction and pipe diameter constructed in step S200, which can clearly show the proximal end and the distal end of the blood vessel.

[0144] The proximal end refers to the end of the blood vessel close to the heart, for example, in the arm arteriovenous fistula, the blood vessel end close to the axillary fossa is the proximal end, and the blood flow direction is usually from the proximal end to the distal end.

[0145] The extension direction refers to the path direction extending from the puncture starting point as the starting point along the proximal end of the blood vessel, which is used as a path guide for subsequent marking of the rotation puncture point.

[0146] In specific implementation, the system calls the target puncture starting point coordinates determined in step S403, combines the blood vessel model constructed in step S200, and analyzes the direction of the blood vessel segment corresponding to the starting point; the proximal end position of the blood vessel segment is identified through the blood vessel model, and the direction from the puncture starting point to the proximal end is determined as the extension direction, for example, extending from the wrist fistula starting point to the elbow.

[0147] Step S404 is based on the target puncture starting point determined in step S403, combines the blood vessel model to determine the extension direction, and provides a direction reference for subsequent orderly marking of the rotation puncture point. It connects the starting point positioning result of S403, determines the extension logic through the blood vessel anatomy structure, avoids the random distribution of rotation points, and ensures that all rotation points are reasonably distributed along the blood vessel.

[0148] S405: Mark a plurality of puncture points as rotation puncture starting points according to the extension direction and a preset reference interval distance;

[0149] The extension direction refers to the direction extending along the proximal end of the blood vessel determined in step S404.

[0150] The preset reference interval distance refers to the minimum distance between adjacent puncture points set based on the clinical rope ladder puncture specification, which is usually 0.5-1.0 cm, to ensure that each puncture point has enough repair time and avoid repeated damage to the blood vessel wall.

[0151] Rotating puncture starting points refer to a plurality of alternative needle entry points marked at a reference interval distance along the extension direction, used for rotation in different puncture periods to avoid repeated puncture at a single location.

[0152] In specific implementation, the system takes the extension direction determined in step S404 as the axis, and marks new puncture points at a preset reference interval distance from the target puncture starting point in step S403, for example, marking a point every 0.8 cm in the proximal end direction from the initial venous end starting point, a total of 4-6 points; after marking, these points collectively form a rotating puncture starting point set, and the three-dimensional coordinates of each point are stored.

[0153] Step S405 marks rotating puncture starting points at a reference interval distance based on the extension direction determined in S404, achieving ordered distribution of puncture points. It is guided by the direction of S404, and ensures that adjacent puncture points are at a reasonable distance through fixed intervals, avoiding both local blood vessel damage accumulation caused by too close spacing and ensuring that all points are distributed along the effective segment of the blood vessel, providing sufficient alternative positions for subsequent puncture rotation and prolonging the service life of the blood vessel access.

[0154] S406: Analyze according to the rotating puncture starting points and the preset puncture point analysis sub-strategy to generate the corresponding puncture angle constraints for each rotating puncture starting point.

[0155] Rotating puncture starting points refer to the plurality of alternative needle entry points marked in step S405.

[0156] The preset puncture point analysis sub-strategy refers to an analysis rule for evaluating the feasibility of the puncture point, including logic for calculating a reasonable puncture angle in combination with parameters such as blood vessel depth, subcutaneous tissue thickness, and muscle texture.

[0157] Puncture angle constraints refer to the range of angles between the puncture needle and the skin surface set for each rotating puncture starting point, ensuring that the puncture needle can accurately reach the target blood vessel.

[0158] In specific implementation, the system retrieves the rotating puncture starting points marked in step S405, combines the blood vessel model and muscle tissue model constructed in step S200, and obtains parameters such as blood vessel depth and subcutaneous tissue thickness corresponding to each starting point; through the preset puncture point analysis sub-strategy, these parameters are substituted into the puncture angle calculation model to generate corresponding angle constraints for each rotating puncture starting point, for example, a certain point has an angle constraint set to 35°-40° due to a deep blood vessel.

[0159] Step S406 generates angle constraints based on the rotation puncture starting point determined in S405 in combination with the puncture point analysis sub-strategy to ensure that the puncture operation specification of each rotation point is feasible. It takes over the rotation point marking result of S405, adapts to the anatomical characteristics of different points through individualized angle constraints, avoids puncture failure or tissue damage caused by improper angle, and at the same time provides clear operation guidance for medical staff, improving the accuracy and safety of rotation puncture.

[0160] With reference to Figure 5 , the path evaluation strategy includes:

[0161] S407: calculating and analyzing with a preset puncture point risk evaluation model to determine the puncture risk weight value of the puncture path of the puncture point corresponding blood vessel;

[0162] The preset puncture point risk evaluation model refers to a model used to quantitatively evaluate puncture risk, which can comprehensively consider the distance of puncture path and risk tissue, blood vessel diameter, puncture angle and other factors. The specific model will be further detailed later.

[0163] The puncture risk weight value refers to the risk quantitative value calculated by the model. The smaller the value, the lower the risk of the puncture path, and the larger the value, the higher the risk.

[0164] Specifically, the system calls the multiple puncture paths generated in step S400, combines the blood vessel model and muscle tissue model constructed in step S200 to obtain the basic data of each path corresponding to the puncture point position, the distance from the key tissue, the blood vessel state, etc.; input these data into the preset puncture point risk evaluation model, and calculate the puncture risk weight value of each path through the model, for example, the risk weight value of a certain path is 0.8 because it is close to the nerve and the angle deviation is large, and the risk weight value of another path is 0.3 because it is far away from the risk tissue and the angle is appropriate.

[0165] Step S407 calculates the risk weight value through the puncture point risk evaluation model based on the multiple puncture paths generated in step S400, realizing the quantitative evaluation of puncture risk. It takes over the path generation result of S400, converts the originally difficult-to-directly-compare paths into quantifiable risk values, provides objective basis for subsequent selection of optimal path, avoids relying on subjective experience to judge risk, and improves the accuracy of risk evaluation.

[0166] S408: arranging the multiple puncture paths in descending order based on the puncture risk weight value to determine the puncture path with the smallest puncture risk weight value and mark it as the optimal puncture path.

[0167] Descending order refers to arranging the puncture risk weight value from large to small to facilitate quick positioning of the path with the lowest risk.

[0168] The optimal puncture path refers to the path with the minimum puncture risk weight value after risk assessment. This path usually has the characteristics of being far away from risk tissues, reasonable angle, and small blood vessel damage.

[0169] In specific implementation, the system collects the puncture risk weight values corresponding to all puncture paths obtained in step S407, and arranges these paths in descending order, for example, arranges the paths with risk weight values 0.8, 0.6, 0.3, and 0.2 in order. From the arrangement result, the path with the minimum risk weight value, for example, the path with the weight value 0.2, is selected and marked as the optimal puncture path, and is displayed in the three-dimensional model with highlighted lines.

[0170] Step S408 selects the optimal puncture path by arranging in descending order based on the puncture risk weight values obtained in step S407, to realize accurate optimization of the puncture path. It locks the path with the lowest risk quickly through sorting based on the risk quantification result of S407, ensures that the finally recommended path has the highest safety, and directly provides operation guidance for medical staff through clear marking, reduces the decision-making time of path selection, and improves the puncture preparation efficiency.

[0171] The puncture point risk assessment model calculates as follows:

[0172]

[0173] wherein, R i is the comprehensive risk index of the i-th puncture point, K is the path type correction coefficient of the preset different path blood vessels, N i is the number of punctures within the last 2 weeks of the i-th puncture point, L max is the maximum length of the effective puncture segment of the blood vessel, L i is the straight-line distance between the i-th puncture point and the anastomosis, D i is the diameter of the blood vessel at the i-th puncture point, which is obtained by ultrasonic measurement, V i is the blood flow velocity at the i-th puncture point, S i is the distance between the i-th puncture point and the adjacent puncture point, α is the preset time attenuation coefficient, T i is the interval time of the i-th puncture point from the last puncture.

[0174] Referring to Figure 6 It also includes a puncture parameter learning step:

[0175] S500: Real-time acquisition of the needle tip position and needle body posture of the physical puncture needle in the three-dimensional image model through electromagnetic tracking or image recognition technology;

[0176] The electromagnetic tracking technology refers to a technology for real-time positioning of the spatial position of an object through electromagnetic induction principle. The needle tip coordinates can be determined through the induction between the electromagnetic sensor on the puncture needle and the external magnetic field generator.

[0177] Image recognition technology refers to the technology of locating the needle tip and the direction of the needle body by analyzing the morphological characteristics of the puncture needle in real-time ultrasound images.

[0178] A physical puncture needle refers to a needle actually used for puncture procedures, including the needle body and needle tip, and is an instrument that directly contacts the patient's tissues during puncture.

[0179] The three-dimensional image model refers to the three-dimensional model of the arm constructed in step S200, which can intuitively display the spatial location of tissues such as blood vessels and muscles.

[0180] The needle tip position refers to the coordinates of the tip of the physical puncture needle in the three-dimensional image model, which is used to determine whether the puncture needle is close to the target blood vessel.

[0181] Needle posture refers to the angle between the physical puncture needle and the skin surface, the direction of the needle, etc., and is used to determine whether the puncture angle meets the preset constraints.

[0182] In practice, the system can choose between electromagnetic tracking or image recognition technology. Electromagnetic tracking requires installing a miniature electromagnetic sensor at the tail of the physical puncture needle. This sensor receives induction signals in real time via an external magnetic field device, calculating the needle tip's coordinates and the needle's posture in the 3D image model. Image recognition, on the other hand, requires processing the real-time ultrasound image to identify the needle's echo lines. The needle's posture is analyzed based on the line's direction, and the needle tip position is located using the line's endpoints. Both technologies enable real-time acquisition of the needle tip position and the needle's posture, which are then simultaneously displayed in the 3D image model.

[0183] Step S500 uses electromagnetic tracking or image recognition technology to acquire the position and orientation of the puncture needle in real time, providing a data foundation for dynamic monitoring of the puncture process. Following the optimal puncture path determined in step S408, it associates the physical puncture operation with the three-dimensional image model, achieving real-time mapping between the physical operation and the digital model, providing a basis for subsequent judgment of path deviation.

[0184] S501: Compare the real-time trajectory of the physical puncture needle with the optimal puncture path. When it deviates from the path or approaches the boundary of the risk tissue and exceeds the preset safety threshold, trigger a visual or auditory warning signal and generate a puncture path offset data packet.

[0185] The real-time trajectory refers to the continuous path of the needle tip movement during the puncture process, which is formed by connecting the needle tip positions obtained in real time in step S500.

[0186] The preset safety threshold refers to the allowable deviation range set based on clinical safety standards, including the path deviation threshold and the risk tissue distance threshold, such as whether the distance to risk tissues such as nerves is greater than the set distance of 3mm.

[0187] Visual or auditory warning signals are signals used to alert the operator. Visual signals may include a flashing red icon on the screen, while auditory signals may include a beeping sound. The greater the deviation, the stronger the signal.

[0188] The puncture path offset data packet refers to a collection of data that records offset-related information, including offset time, offset distance, and the type of organization approaching the risk.

[0189] In practice, the system compares the real-time trajectory obtained in step S500 with the optimal puncture path determined in step S408 point by point and calculates the distance deviation between the two; at the same time, it monitors the distance between the needle tip and the boundary of the risk tissue in real time.

[0190] If the trajectory deviates from the optimal path by 3mm at a certain moment, exceeding the path deviation threshold of 2mm, or if the distance between the needle tip and the nerve is only 2mm, which is lower than the risk tissue distance threshold of 3mm, the system will immediately trigger a visual light warning and a set sound prompt warning, and automatically record the time, deviation value and other information of this deviation, and generate a puncture path deviation data packet.

[0191] Step S501 compares the real-time trajectory obtained in step S500 with the optimal puncture path and triggers an early warning, thereby achieving safe monitoring of the puncture process. It inherits the real-time monitoring data from S500, judges the safety of the operation through preset safety thresholds, promptly alerts the operator when risks occur, and generates offset data packets to record abnormal situations. This reduces the risk of puncture errors and provides raw data for subsequent analysis.

[0192] S502: Based on the input of the puncture path offset data packet, the preset automatic puncture path planning agent performs path offset analysis to generate puncture offset feedback results.

[0193] An automated puncture path planning agent refers to an algorithmic model with analytical and learning capabilities. It can summarize deviation patterns through historical deviation data and determine the causes of deviation, such as deviations in operating angles or the influence of tissue elasticity.

[0194] Path offset analysis refers to interpreting information in puncture path offset data packets, including the degree of offset, the puncture stage in which the offset occurred, and whether it is related to a specific tissue type.

[0195] The puncture deviation feedback result refers to the conclusion generated after analysis by the intelligent agent, including the cause of the deviation, correction suggestions, such as adjusting the puncture angle by 10°, and subsequent puncture precautions, such as needing to slowly insert the needle in a hard area.

[0196] In practice, the system inputs the puncture path offset data packet generated in step S501 into a preset automatic puncture path planning agent. The agent calls its built-in analysis logic to analyze information such as the offset distance, timing of occurrence, and risk tissue type in the data packet. For example, if multiple offsets are found to occur at a blood vessel depth of 8mm and in the same direction, it is determined that the thick subcutaneous tissue in that area may be causing the needle insertion angle deviation. Based on the analysis results, a puncture offset feedback result is generated, clarifying the cause of the offset and providing correction suggestions for adjusting the needle insertion angle to the set angle.

[0197] Step S502, based on the puncture path offset data packet generated in step S501, generates feedback results through intelligent agent analysis, achieving closed-loop optimization of the puncture operation. It inherits the offset data record from S501 and transforms the original data into actionable correction suggestions through intelligent analysis. This helps the operator adjust the puncture action in a timely manner and provides an optimization basis for subsequent puncture path planning, continuously improving puncture accuracy.

[0198] Reference Figure 7 The scanning strategy includes:

[0199] S101: Perform limb analysis based on the patient's preset puncture target area to determine the puncture limb type and match the ultrasound scanner movement speed corresponding to the puncture limb type in the preset ultrasound scanning database.

[0200] The type of limb to be punctured refers to the limb and specific location on which the patient is punctured. The depth of blood vessels and the density of tissue vary in different locations such as the forearm and upper arm.

[0201] The preset ultrasound scan database refers to a database that stores the scanning parameters corresponding to different limb types. It includes the standard ultrasound scanner movement speed for parts such as the forearm and upper arm. For example, the scanning speed for the forearm is 0.5-1cm / s, and the scanning speed for the upper arm is 0.3-0.8cm / s due to the thicker tissue.

[0202] The ultrasound scanner's moving speed refers to the speed at which the ultrasound probe moves across the surface of a limb. The speed must be adapted to the type of limb to ensure clear images, avoiding both excessively fast movements that result in blurry images and excessively slow movements that affect efficiency.

[0203] In specific execution, the system calls the preset puncture target area determined in step S100, such as the fistula area of ​​the patient's left forearm, performs limb analysis on the area, and determines that the puncture limb type is the mid-forearm; retrieves the corresponding ultrasound scanner movement speed from the preset ultrasound scanning database, and matches it to a speed range of 0.5-1cm / s; sets this speed as the reference speed for the current scan.

[0204] Step S101 determines the limb type to be punctured based on the preset puncture target area and matches the scanning speed, providing suitable speed parameters for subsequent scans. It follows the target area setting in step S100 of ultrasound image acquisition. By accurately matching the limb type with the speed, it avoids image quality problems caused by universal speeds, laying the foundation for obtaining clear ultrasound image data.

[0205] S102: Perform ultrasound scanning on the preset puncture target area based on the moving speed of the ultrasound scanner, and analyze the real-time fit of the preset ultrasound scanner, keeping the real-time fit within the preset reference fit range.

[0206] Real-time fit refers to the degree of contact between the ultrasound probe and the skin surface. It can be evaluated by pressure sensors or image signal intensity. For example, the closer the contact, the less the ultrasound signal attenuates and the more uniform the image grayscale.

[0207] The preset baseline fit range refers to the probe fit standard that ensures the quality of ultrasound images. It is usually judged based on a signal intensity greater than or equal to 80% and the absence of obvious air artifacts in the image.

[0208] In practice, the system controls the ultrasound probe to move and scan along the preset puncture target area according to the ultrasound scanner movement speed determined in step S101. Simultaneously, the system monitors the real-time fit using a pressure sensor built into the probe, and comprehensively evaluates the fit based on the signal intensity and artifact status of the ultrasound image. If the signal intensity is 90% and there are no air artifacts, the fit meets the benchmark range; if the signal intensity is 60% and artifacts appear due to limb wrinkles, the fit does not meet the standard. The system automatically adjusts the probe pressure, such as by slightly increasing the pressure, to maintain the real-time fit within the benchmark range.

[0209] Step S102 performs a scan based on the scanning speed determined in S101 and monitors the real-time fit to ensure the stability of the scanning process. Following the speed setting in S101, it compensates for differences in limb surface morphology, such as wrinkles and protrusions, through fit control, avoiding image loss due to poor fit and ensuring that the acquired ultrasound image data is complete and clear, providing reliable raw data for the subsequent model construction step S200.

[0210] S103: When the ultrasound scanner’s moving speed and fit do not meet the preset scanning conditions, a moving speed prompt or fit adjustment prompt is issued.

[0211] The preset scanning conditions refer to the comprehensive standards that ensure scanning quality, including the ultrasonic scanner's moving speed being within the speed range matched by S101 and the real-time fit being within the reference fit range set by S102.

[0212] The movement speed prompt command refers to the prompt issued when the scanning speed deviates from the matching range. For example, if the speed is too fast, a text prompt will be displayed asking you to reduce the movement speed to 0.5-1cm / s.

[0213] The fit adjustment prompt command refers to the prompt issued when the fit is not up to standard. For example, if the fit is too loose, a voice prompt will be issued to increase the probe pressure.

[0214] During execution, the system monitors the ultrasound scanner's movement speed and real-time fit in real time. If the speed is detected to rise to 1.2 cm / s, exceeding the 1 cm / s upper limit for forearm type matching, a movement speed warning is triggered, displaying red text on the operation interface indicating that the speed is too fast and requesting a reduction. If the fit signal strength is detected to drop to 70%, below the baseline of 80%, a fit adjustment warning is triggered, issuing a voice prompt indicating poor fit and requesting adjustment of the probe position. After the operator adjusts according to the prompts, the system re-monitors until the preset scanning conditions are met.

[0215] Step S103 issues a prompt command when the scanning speed or fit does not meet the requirements, realizing dynamic correction of the scanning process. It follows the real-time monitoring results of S102, and avoids the generation of unqualified scanning data by timely prompts, ensuring the stability of the quality of the finally acquired ultrasound image data, and providing reliable basic data support for subsequent model construction and path planning.

[0216] Based on the same inventive concept, embodiments of the present invention provide an ultrasound-guided intelligent puncture path planning system, comprising:

[0217] The ultrasound image acquisition module performs ultrasound scanning on the patient's preset puncture target area using a preset scanning strategy to obtain ultrasound image data, and simultaneously collects the arterial and venous blood flow in the preset puncture target area.

[0218] The puncture model construction module constructs a three-dimensional model of the arm corresponding to the patient's puncture target area based on ultrasound image data, and analyzes the three-dimensional model of the arm through a feature recognition algorithm to determine the vascular model and muscle tissue model;

[0219] The risk avoidance module analyzes the vascular model and arterial and venous blood flow to identify target blood vessels and key tissues that meet the puncture conditions, and constructs a three-dimensional safe puncture space within a preset safe distance threshold based on the identification results of target blood vessels and key tissues.

[0220] The puncture path generation module generates multiple puncture paths within a three-dimensional safe puncture space that meet the puncture angle constraints based on the target blood vessels and key tissues, and is configured with a path evaluation strategy to evaluate and select the optimal puncture path.

[0221] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0222] This invention provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as an ultrasound-guided intelligent puncture path planning method.

[0223] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.

[0224] Based on the same inventive concept, embodiments of the present invention provide a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as an ultrasound-guided smart puncture path planning method.

[0225] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0226] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification, including the abstract and drawings, can be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is merely one example of a series of equivalent or similar features.

Claims

1. A method for intelligent puncture path planning under ultrasound guidance, characterized in that, include: The ultrasound image acquisition step involves using a preset scanning strategy to perform ultrasound scanning on the patient's preset puncture target area to obtain ultrasound image data, and simultaneously collecting the arterial and venous blood flow in the preset puncture target area. The steps for constructing the puncture model are as follows: a three-dimensional model of the arm corresponding to the patient's puncture target area is constructed based on ultrasound image data, and the three-dimensional model of the arm is analyzed through feature recognition algorithms to determine the vascular model and muscle tissue model. The risk avoidance steps involve analyzing vascular models and arterial and venous blood flow to identify target vessels and key tissues that meet the puncture conditions, and constructing a three-dimensional safe puncture space within a preset safe distance threshold based on the identification results of target vessels and key tissues. The puncture path generation step generates multiple puncture paths within a three-dimensional safe puncture space that meet the puncture angle constraints based on the target blood vessels and key tissues, and is equipped with a path evaluation strategy to evaluate and select the optimal puncture path.

2. The ultrasound-guided intelligent puncture path planning method according to claim 1, characterized in that, The risk organization avoidance steps also include a risk disease identification strategy, including: The identification and analysis based on the vascular model are used to determine the rate of change of arterial and venous vessel diameter, and the results of the comparison of the rate of change of standard vessel diameter corresponding to the puncture target area in the preset vascular disease database are obtained. Based on the muscle tissue model, identification and analysis are performed to obtain the muscle tissue model parameters, and the parameters are compared with a pre-set muscle disease model database to determine the comparison results of the muscle tissue models. Based on the comparison results of changes in blood vessel diameter and the comparison results of muscle models, the locations of blood vessels and muscle regions that do not meet the preset puncture risk conditions are marked to determine the puncture avoidance area. The overlapping areas are divided based on the three-dimensional safe puncture space and the puncture avoidance area to determine the three-dimensional safe puncture space of the puncture avoidance area.

3. The ultrasound-guided intelligent puncture path planning method according to claim 1, characterized in that, The puncture path generation step also includes a puncture point analysis sub-strategy: Based on the analysis of ultrasound image parameters, the skin surface tissue image of the puncture target area is obtained, and image feature recognition is performed to determine whether the skin surface is consistent with the preset skin damage characteristics; Areas that match the preset skin damage characteristics are marked to determine non-puncture initiation point areas and available puncture initiation point areas; Based on the available puncture starting point area as the puncture path, an associated region is generated, and the target puncture starting point is obtained according to the optimal puncture path marking. The target puncture starting point includes the venous end starting point and the arterial end starting point.

4. The ultrasound-guided intelligent puncture path planning method according to claim 3, characterized in that, It also includes a puncture angle constraint optimization strategy for the target puncture starting point: The aforementioned puncture angle constraint strategy calculates the optimal puncture angle using a preset puncture constraint angle model. The puncture constraint angle model is calculated using the following formula: Where, θ optimal The optimal puncture angle is the angle between the puncture point and the skin surface, where d is the depth of the target blood vessel measured by ultrasound, and ρ is the puncture depth. m φ is the preset muscle texture resistance coefficient of the muscle in the target puncture initiation point area when the puncture needle is inserted, φ is the angle between the blood vessel direction and the skin surface as measured by ultrasound, t is the subcutaneous tissue thickness as measured by ultrasound, β is the muscle tension adjustment coefficient, and T is the muscle texture resistance coefficient. m This is the preset muscle tension index.

5. The ultrasound-guided intelligent puncture path planning method according to claim 3, characterized in that, It also includes a strategy for constructing rotational puncture initiation points: Based on the puncture starting point and the vascular model, the proximal end of the vascular model is determined as the direction of extension. Multiple puncture points are marked as rotation puncture starting points according to the extension direction and the preset reference interval distance; The analysis is performed based on the rotation puncture start point and the preset puncture point analysis sub-strategy to generate the corresponding puncture angle constraint for each rotation puncture start point.

6. The ultrasound-guided intelligent puncture path planning method according to claim 1, characterized in that, The path evaluation strategy includes: The risk weight value of puncture point is determined by calculating and analyzing the pre-set puncture point risk assessment model to determine the puncture risk weight value of the corresponding blood vessel when puncture is performed according to the puncture path. Multiple puncture paths are sorted in descending order based on puncture risk weight values ​​to determine the puncture path with the smallest puncture risk weight value and mark it as the optimal puncture path.

7. The ultrasound-guided intelligent puncture path planning method according to claim 6, characterized in that, The puncture point risk assessment model is calculated using the following formula: Among them, R i Let N be the comprehensive risk index for the i-th puncture point, K be the correction coefficient for different pathway vessel types, and N be the total risk index for the puncture point. i L represents the number of punctures performed at the i-th puncture point within the past two weeks. max L is the maximum effective puncture segment length for vascular access. i Let D be the straight-line distance between the i-th puncture point and the anastomosis. i V is the diameter of the blood vessel at the i-th puncture point, obtained by ultrasound measurement. i Let S be the blood flow velocity at the i-th puncture point. i Let T be the distance between the i-th puncture point and its adjacent puncture points, α be the preset time decay coefficient, and T be the distance between the i-th puncture point and its adjacent puncture points. i Let be the time interval between the i-th puncture point and the last puncture.

8. The ultrasound-guided intelligent puncture path planning method according to claim 1, characterized in that, It also includes the puncture parameter learning steps: The needle tip position and needle body posture of the physical puncture needle in the three-dimensional image model are obtained in real time by electromagnetic tracking or image recognition technology. The real-time trajectory of the physical puncture needle is compared with the optimal puncture path. When it deviates from the path or approaches the boundary of the risk tissue and exceeds the preset safety threshold, a visual or auditory warning signal is triggered and a puncture path offset data packet is generated. Based on the puncture path offset data packet input, a pre-set automatic puncture path planning agent performs path offset analysis to generate puncture offset feedback results.

9. The ultrasound-guided intelligent puncture path planning method according to claim 1, characterized in that, The scanning strategy includes: Based on the patient's preset puncture target area, limb analysis is performed to determine the type of puncture limb and match the ultrasound scanner movement speed corresponding to the type of puncture limb in the preset ultrasound scanning database. Based on the moving speed of the ultrasound scanner, the preset puncture target area is scanned with ultrasound, and the real-time fit of the preset ultrasound scanner is analyzed to keep the real-time fit within the preset reference fit range. When the ultrasound scanner's movement speed and fit do not meet the preset scanning conditions, a movement speed prompt or fit adjustment prompt will be issued.

10. An ultrasound-guided intelligent puncture path planning system, employing the method described in any one of claims 1-9, characterized in that, include: The ultrasound image acquisition module performs ultrasound scanning on the patient's preset puncture target area using a preset scanning strategy to obtain ultrasound image data, and simultaneously collects the arterial and venous blood flow in the preset puncture target area. The puncture model construction module constructs a three-dimensional model of the arm corresponding to the patient's puncture target area based on ultrasound image data, and analyzes the three-dimensional model of the arm through a feature recognition algorithm to determine the vascular model and muscle tissue model; The risk avoidance module analyzes the vascular model and arterial and venous blood flow to identify target blood vessels and key tissues that meet the puncture conditions, and constructs a three-dimensional safe puncture space within a preset safe distance threshold based on the identification results of target blood vessels and key tissues. The puncture path generation module generates multiple puncture paths within a three-dimensional safe puncture space that meet the puncture angle constraints based on the target blood vessels and key tissues, and is configured with a path evaluation strategy to evaluate and select the optimal puncture path.

Images